Current estimates indicate that as many as 100 million people in over 70 countries are drinking water with arsenic (As) concentrations up to 100 times the World Health Organization (WHO) guideline of 10 ug per liter. There is significant variability in progression from As exposure to clinical manifestations of disease, and it is thought that genetic and nutritional factors may together account for a substantial portion of this variability. Thus, this proposal seeks to develop biomarkers that predict risk for the development of arsenicosis. Ingested inorganic As (InAs) is methylated to methylarsonic (MMA) and dimethylarsinic (DMA) acids via folate- dependent one-carbon metabolism. A reduced capacity to fully methylate As to DMA is associated with reduced urinary elimination of As. DMAV is less toxic and has a much shorter circulating half-live than InAs, and it is rapidly excreted in urine. There is great inter-individual variability in As methylation capacity, and epidemiological evidence, including our own, suggests that people who are good methylators are at reduced risk for As-induced skin lesions, cancers and cardiovascular disease. However, a growing body of evidence from in vitro and laboratory animal studies indicates that the trivalent form of MMA, a requisite intermediate in the pathway toward DMA synthesis, is extremely toxic. Our work over the past several years in Bangladesh has made seminal findings regarding the strong impact of nutritional regulation of one-carbon metabolism on the inter-individual variability in As methylation, blood As levels, and risk for arsenic-induced skin lesions, e.g., our clinical trial demonstrated that folic acid supplementation lowered blood As and blood MMA concentrations. Our 1st aim is to conduct a nested case-control study of 1,000 incident premalignant skin lesion cases and 1,000 controls to test the hypotheses that the capacity to methylate As is influenced by As methyltransferase (AS3MT) genotypes and that AS3MT genotypes and As methylation are associated with risk for skin lesions. Our 2nd aim will examine whether polymorphisms in one-carbon metabolism candidate genes are associated with hyperhomocysteinemia (which is very common in this population, and is associated with reduced capacity to methylate As to DMA), with methylation of As, methylation of leukocyte DNA, and with risk for skin lesions. We will also examine whether these associations are influenced by folate status. Collectively, these aims will allow us to identify of a set of biomarkers (As metabolites, leukocyte DNA methylation, folate, B12, Hcys, and SNPs) to identify sub-groups of individuals who may be at increased risk for As-induced pathology utilizing biological specimens that are already in hand. These studies will also resolve the critical issue of whether or not a higher capacity to methylate As is beneficial. Results of these studies will create opportunities for evaluation of interventions among high risk groups. They will also result in the identification of biomarkers to facilitate targeted therapies that exploit the mechanisms involved in susceptibility, resulting in low-risk, low-cost therapeutic strategies that could potentially reduce disease risk for hundreds of thousands of people.